Molten silver and copper are completely soluble in each other in all proportions. However, alloys which have copper contents ranging from about 2% through 27%, when solidified and examined under a microscope, exhibit two discrete constituents: one is nearly 100% silver; the other is a silver-copper “eutectic” (71.9% silver; 28.1% copper), whose melting point is 1435° F. (780° C.). When standard sterling silver is cooled, microscopic analysis shows both of the above constituents to be present in the solidified sterling. The alloy is entirely liquid at 1640° F. (890° C.) and entirely solid at 1435° F. (780° C.) However, the degree of copper solubility in the solid alloy depends on the heat treatment used, and the overall physical properties of the sterling can be materially affected, not only by heating the silver to different temperatures, but also by employing different cooling rates.
Silver alloys are normally supplied soft for easy working. Heat treatment can be used to increase hardness (and decrease ductility). The process, known as precipitation hardening involves heating and cooling the silver in such a way as to cause copper to precipitate out of solid solution, thereby producing a fine binary structure. This type of structure is hard, but it is also difficult to work, and has a tendency to crack. Precipitation hardening of conventional sterling silver can be achieved by (a) heating the alloy to or above 775° C., (b) holding the alloy at that temperature for 15-30 minutes for annealing thereof (i.e. dissolving all the copper in the silver), (c) quenching rapidly in cold water, which prevents formation of Cu-rich coarse precipitates which are ineffective in bringing about hardening, (d) re-hardening the softened alloy by heating to e.g. 300° C. for 30-60 minutes resulting in the formation of very tine Cu-rich particles which are effective in bringing about hardening and (e) air cooling. The annealing temperatures involved are very high and are close to the onset of melting. Furthermore, there are very few times in practical production that a silversmith can safely quench a piece of nearly finished work because of the risk of distortion of the article being made and/or damage to soldered joints. Silversmiths therefore regard precipitation hardening of sterling silver as of metallurical interest only. It is too difficult for commercial or industrial production of articles of jewelry, silver plate, hollowware, and the like (see Fischer-Buhner, “An Update on Hardening of Sterling Silver Alloys by Heat Treatment”, Proceedings, Santa Fe Symposium on Jewellery Manufacturing Technology, 2003, 20-47 at p. 29) and it is unnecessary because sterling silver as produced generally has hardness of 70 Vickers or above. Alloys of higher Vickers hardness are obtained by work hardening rather than precipitation hardening. Furthermore, in investment casting, it is impractical for the manufacture of silver rings and other jewelry by stone in place casting where there is a significant risk that the stone will not survive annealing and quenching.
It has been alleged by Aldo Reti, Encyclopaedai of Materials, Science and Technology that:
“As cast, sterling silver (7.5 wt % Cu) has a silver-rich primary solid solution and some non-equilibrium eutectic. As wrought, with conventional 590-620° C. anneals, it has a silver matrix (containing little copper) with laminated copper particles (containing little silver). Sterling silver may be precipitation hardened by dissolving the copper at 745° C. and precipitating it at 310° C. and it may be hardened appreciably by annealing at 695-745° C. and air cooling. If a handle is torch brazed to a pitcher and the pitcher is air cooled the annealed brazed areas will harden; if a fork is annealed at 670-700° C. it will harden on air cooling. Air cooling is convenient and sterling silver is very soft in a furnace at its solution temperature.”
In the example given by Reti, the pitcher and handle would have been torch annealed between the forming/shaping processes. A brazing test was therefore conducted by the inventor on traditional sterling silver sheet (7.5 wt % Cu) that had been initially torch annealed to replicate Reti's example. Soldering to the sheet was with an easy solder/brazing alloy melting range 705-720° C. and flux was placed on a pickled surface of the sheet prior to soldering which was carried out using a gas/air torch. Hardness levels were measured on a Vickers Hardness machine at CATRA (Cutlery and Allied Trades Research Association) in Sheffield, England. Results were as follows:
Hardness MeasurementsAverageHV 2½ kghardnessSample12345HV1.67.066.567.566.564.566.42.59.657.959.160.458.759.13.61.865.065.065.064.164.21. As-received (annealed by manufacturer);2. Torch annealed and air cooled;3. Torch annealed, air cooled, brazed with Easy soldering/brazing alloy and air cooled.
The sheet that had been torch annealed and air cooled had a slightly greater hardness than the sheet annealed by the manufacturer. There was no appreciable hardening response of the torch annealed traditional sterling silver sheet after brazing—the measured hardness did not exceed that of commercially available sheet in the annealed condition. The precipitation hardening alleged by Reti was not observed, and the present result is in accordance with common general knowledge amongst silversmiths i.e. that standard sterling does not precipitation harden under these conditions.
Patent GB-B-2255348 (Rateau, Albert and Johns; Metaleurop Recherche) discloses a novel silver alloy that maintains the properties of hardness and lustre inherent in Ag—Cu alloys while reducing problems resulting from the tendency of the copper content to oxidise. The alloys are ternary Ag—Cu—Ge alloys containing at least 92.5 wt % Ag. 0.5-3 wt % Ge and the balance, apart from impurities, copper. The alloys are stainless in ambient air during conventional production, transformation and finishing operations, are easily deformable when cold, easily brazed and do not give rise to significant shrinkage on casting. They also exhibit superior ductility and tensile strength. Germanium was stated to exert a protective function that was responsible for the advantageous combination of properties exhibited by the new alloys, and was in solid solution in both the silver and the copper phases. The microstructure of the alloy was said to be constituted by two phases, a solid solution of germanium and copper in silver surrounded by a filamentous solid solution of germanium and silver in copper which itself contains a few intermetallic CuGe phase dispersoids. The germanium in the copper-rich phase was said to inhibit surface oxidation of that phase by forming a thin GeO and/or GeO2 protective coating which prevented the appearance of firestain during brazing and flame annealing. Furthermore the development of tarnish was appreciably delayed by the addition of germanium, the surface turned slightly yellow rather than black and tarnish products were easily removed by ordinary tap water. It is explained that increased hardness can be developed by de-tensioning the alloy e.g. by heating to 500° C. and then heating the alloy to a “low annealing” temperature below 400° C. e.g. to 200° for 2 hours to give a Vickers hardness of about 140. However, there is no suggestion that such hardness can be achieved without the steps of heating to an annealing temperature followed by quenching, and therefore there is also no suggestion that the increased hardness can be achieved in nearly finished work.
U.S. Pat. No. 6,168,071 and EP-B-0729398 (Johns) disclosed a silver/germanium alloy which comprised a silver content of at least 77 wt % and a germanium content of between 0.4 and 7%, the remainder principally being copper apart from any impurities, which alloy contained elemental boron as a grain refiner at a concentration of greater than 0 ppm and less than 20 ppm. The boron content of the alloy could be achieved by providing the boron in a master copper/boron alloy having 2 wt % elemental boron. It was reported that such low concentrations of boron surprisingly provided excellent grain refining in a silver/germanium alloy, imparting greater strength and ductility to the alloy compared with a silver/germanium alloy without boron. The boron in the alloy inhibited grain growth even at temperatures used in the jewelry trade for soldering, and samples of the alloy were reported to have resisted pitting even upon heating repeatedly to temperatures where in conventional alloys the copper/germanium eutectic in the alloy would melt. Strong and aesthetically pleasing, joints between separate elements of the alloy could be obtained without using a filler material between the free surfaces of the two elements and a butt or lap joint could be formed by a diffusion process or resistance or laser welding techniques. Compared to a weld in Sterling silver, a weld in the above-described alloy had a much smaller average grain size that improved the formability and ductility of the welds, and an 830 alloy had been welded by plasma welding and polished without the need for grinding. Again there is no disclosure or suggestion that precipitation hardening can be achieved safely in nearly finished work.
Argentium (Trade Mark) sterling comprises Ag 92.5 wt % and Ge 1.2 wt %, the balance being copper and about 4 ppm boron as grain refiner. The Society of American Silversmiths maintains a website for commercial embodiments of the above-mentioned alloys known as Argentium (Trade Mark) at the web address hup://www.silversmithing.com/largentium.htm. it discloses that Argentium Sterling is precipitation hardenable (i.e. by heating to an annealing temperature and quenching), that a doubling in final hardness can be achieved by reheating at temperatures obtainable in a domestic oven e.g. 450° F. (232° C.) for about 2 hours or 570° F. (299° C.) for about 30 minutes. It further discloses that the hard alloy can be softened by conventional annealing (i.e. heating to an annealing temperature and quenching) and then hardened again if required. However, there is no suggestion that precipitation hardening is appropriate for nearly finished work and that the problems of distortion and damage to soldered joints can be avoided.
U.S. Pat. No. 6,726,877 (Eccles) discloses inter alia an allegedly fire scale resistant, work hardenable jewellery silver alloy composition comprising 81-95.409 wt % Ag. 0.5-6 wt % Cu, 0.05-5 wt % Zn, 0.02-2 wt % Si, 0.01-2 wt % by weight B. 0.01-1.5 wt % In and 0.01-no more than 2.0 wt % Ge. The germanium content is alleged to result in alloys having work hardening characteristics of a kind exhibited by conventional 0.925 silver alloys, together with the firestain resistance of allegedly firestain resistant alloys known prior to June 1994. Amounts of Ge in the alloy of from about 0.04 to 2.0 wt % are alleged to provide modified work hardening properties relative to alloys of the firestain resistant kind not including germanium, but the hardening performance is not. linear with increasing germanium nor is the hardening linear with degree of work. The Zn content of the alloy has a bearing on the colour of the alloy as well as functioning as a reducing agent for silver and copper oxides and is preferably 2.0-4.0 wt %. The Si content of the alloy is preferably adjusted relative to the proportion of Zn used and is preferably 0.15 to 0.2 wt %. Precipitation hardening following annealing, is not disclosed, and there is no disclosure or suggestion that the problems of distortion and damage to soldered joints in nearly finished work made of this alloy can be avoided.
By way of background, U.S. Pat. No. 4,810,308 (Leach & Garner) discloses a hardenable silver alloy comprising not less than 90% silver; not less than 2.0% copper; and at least one metal selected from the group consisting of lithium, tin and antimony. The silver alloy can also contain up to 0.5% by weight of bismuth. Preferably, the metals comprising the alloy are combined and heated to a temperature not less than 1250-1400° F. (676-760° C.) e.g. for about 2 hours to anneal the alloy into a solid solution, a. temperature of 1350° (732° C. being used in the Examples. The annealed alloy is then. quickly cooled to ambient temperature by quenching. it can then be age hardened. by reheating to 300-700° F. (149-371° C.) for a predetermined time followed by cooling of the age hardened alloy to ambient temperature. The age-hardened alloy demonstrates hardness substantially greater that that of traditional sterling silver, typically 100 HVN (Vickers Hardness Number), and can being returned by elevated temperatures to a relatively soft state. The disclosure of U.S. Pat. No. 4,869,757 (Leach & Garner) is similar, In both cases the disclosed annealing temperature is higher than that of Argentium, and neither reference discloses firestain or tarnish-resistant alloys. The inventor is not aware of the process disclosed in. these patents being used for commercial production, and again there is no disclosure or suggestion that hardening can be achieved in nearly finished work.
A silver alloy called Steralite is said to be covered by U.S. Pat. Nos. 0,581,7195 and 5.882,441 and to exhibit high tarnish and corrosion resistance, The alloy of U.S. Pat. No. 5,817.195 (Davitz) contains 90-92.5 wt % Ag, 5,75-5.5 wt % Zn, 0.25 to less than 1 wt % Cu, 0.25-0.5 wt % Ni, 0.1-0.25 wt % Si and 0.0-0.5 wt % In. The alloy of U.S. Pat. No. 5,882,441 (Davitz) contains 90-94 wt % Ag, 3.5-7.35 wt % Zn, 1-3 wt % Cu and 0.1-2.5 wt % Si. A similar high zinc low copper alloy is disclosed in U.S. Pat. No. 4,973,446 (Bernhard) and is said to exhibit reduced firestain, reduced porosity and reduced grain scale. None of these references discusses annealing or precipitation hardening.